Tourists' satisfaction and sense of belonging in adopting responsible behaviors: the role of on-site and social media involvement in cultural tourism

The recovery of tourist flows post-Covid is reaffirming the significant growth trend that has characterized cultural tourism over the last decade. Pre-pandemic tourism data from the UNWTO (2019) revealed that cultural tourism represented approximately 40% of overall tourism worldwide, and the prospects of the industry see a steady growth in the future.

This important growth of cultural tourism has attracted greater interest as evidenced by academic studies (Du Cros and McKercher, 2020; Richards, 2018) in areas such as cultural consumption (Conti et al., 2020; Domínguez-Quintero et al., 2019), cultural motivation (Carreira et al., 2022), heritage conservation (Hayat and Ibrahim, 2022; Vesci et al., 2021; Bui et al., 2020), cultural experiences (Mastroberardino et al., 2022; Palumbo et al., 2022), sustainability topics (De Oliveira et al., 2022; Park et al., 2019) and in consideration of the significant impact that this segment is having and will have on touristic destinations (Spencer and Sargeant, 2022; Alazaizeh et al., 2019).

Moreover, the growing focus on cultural tourism is also explained in light of the 2030 agenda's goals, particularly concerning the current debate about the relationship between cultural, social, environmental sustainability and tourism consumers' behaviors at the community and destination level (Alonso-Muñoz et al., 2023).

However, despite the relevance of this topic, its analysis seems unexplored, especially for what concerns the investigation of the main antecedents influencing tourists' attitudes and behaviors toward the destination (e.g. Hung et al., 2019). Indeed, even if different authors have particularly analyzed the influence of tourists' satisfaction on their behaviors and attitudes (Shrivastav, 2023; Vojtko et al., 2022; Su et al., 2018; Lee et al., 2012), few studies have focused their attention on the possible role of sense of belonging (SOB). This construct represents a more intimate and long-lasting bond between destination and visitor (Jones et al., 2000) because it refers to an actual identifiable sentiment or attachment felt, by tourists, vis à vis the cultural destination that they have visited (Hung et al., 2019; Lin et al., 2014). In particular, SOB can lead tourists to act in favor of the destination brand and of the community itself by also influencing their environmentally responsible (Alonso-Vasquez et al., 2019; Kumar and Nayak, 2019) and loyalty behaviors toward the destination, such as favorable word-of-mouth or recommendation intention (Han et al., 2019a; Prayag et al., 2017; Martin et al., 2013; Tuškej et al., 2013). In this respect, it becomes extremely important to analyze SOB, especially in relation to cultural destinations since when touristic activities are focused on heritage and culture, the tourist-destination relationship is further accentuated (Stebbins, 1997). In many cases, it is the tourists themselves who show an interest in developing a SOB to the cultural site (Poria et al., 2004).

Within this scenario, literature underlines an interesting research gap related to the necessity of deepening the analysis of the SOB influence on the tourist's intentions and behaviors in cultural destinations (Hung et al., 2019). Notably, although several studies have already attempted to analyze the SOB influence in non-tourist (e.g. Natarajan and Veera Raghavan, 2023; Pinna et al., 2018) and tourist sectors (e.g. Mendes et al., 2020; Hahm et al., 2018; Seyitoğlu and Çevik, 2016), it is only recently that attention has been drawn to specific contexts like cultural heritage (Sestino et al., 2023; Price and Applebaum, 2022; Roque, 2022; Hung et al., 2019). Moreover, SOB has been analyzed mostly in the context of brand/product (e.g. Van Tonder and Petzer, 2021), rather than in that of services and experiences (Hung et al., 2019). Thirdly, extant literature emphasizes a further challenging gap related to the necessity of identifying the main factors leading to the SOB creation (Farzinfar et al., 2023). More in detail, although existing contributions have analyzed the influence of satisfaction on SOB (e.g. Peng and Chen, 2020), it emerges a paucity of studies attempting to investigate the role of the tourists' onsite involvement within this relationship (Hung et al., 2019), especially for what concerns its social media dimension (Splendiani et al., 2023). Indeed, the majority of existing research has focused attention on the social media role in the planning phase of the holiday (Pop et al., 2022; Tuclea et al., 2020) without deepening the analysis of their immersive functions in the travel one (Cassia et al., 2021).

Based on these premises, the paper aims to investigate the relationship between satisfaction and SOB in the context of cultural destinations by especially investigating the possible moderating role of on-site involvement relative to both on-site activity and online involvement on social media platforms. Secondly, it analyzes the SOB behavioral outcomes by specifically focusing on the exploration of the relationship between SOB and the adoption of tourists' Environmentally Responsible Behavior (ERB) during their stay at the destination, with further analysis of whether this behavior can have positive spillover effects (Wu et al., 2021) on future attitudes/behaviors (i.e. pro-environmental behavior and recommendation intention), which would foster a virtuous process in tourists' formation as responsible and loyal travelers. In particular, the analysis of this connection appears central to the theme surrounding the quality sustainable development of destinations (Wang et al., 2022; Jopp et al., 2022; Dwyer et al., 2019) and, in particular, of the cultural ones (Spencer and Sargeant, 2022).

By doing so, the paper attempts to fill relevant research gaps related to the necessity of deepening the analysis of (1) the antecedents influencing tourists' loyalty and environmentally responsible attitudes and behaviors, especially by investigating the role of SOB (Hung et al., 2019); (2) the factors affecting the SOB creation through the investigation of the moderating role of onsite involvement (i.e. on-site activity involvement and social media involvement) in the relationship between SOB and satisfaction.

Managerially, the study provides relevant guidelines related to the enhancement of the tourists' SOB and responsible behaviors toward the destinations for policy makers, tourism operators and Destination Marketing Organizations (DMOs), which are currently called to contribute in building on-site and online long-term relationships with tourists (Leung and Jiang, 2018), especially through the implementation of marketing and communication strategies based on the dissemination of sustainable practices (Băcilă et al., 2022).

The remainder of the paper is structured as follows: section two offers a literature review to present the theoretical background and the research hypotheses; section three is devoted to the methodology and the findings are presented in section four. Section five concludes the paper with a discussion of the findings, followed by conclusions, implications and suggested avenues for future research.

In the tourism literature, the terms “sense of belonging”, “destination belongingness” and “place belonging” (Hammitt et al., 2004; Jones et al., 2000) have often been used interchangeably to describe the bond felt by tourists and a destination (Kumar and Nayak, 2019). According to Cheng and Kuo (2015, p. 547), a SOB to a destination can be defined as “a sense of affiliation with or feeling a sense of membership with an environment”. It can be interpreted as an overall emotional attachment, in which people develop a strong relational bond with a place or an environment (Seamon, 1979). This attachment can be created through lived experiences (Pine and Gilmore, 1999), which should be carefully staged by the operators present within the destination. Notably, among them, DMOs assume today a key role in building these experiences in the guise of tourist show directors (Karayilan and Cetin, 2016).

Moreover, in the perspective put forward by Jones et al. (2000), SOB also denotes a bond that emerges from within an individual and thus can be felt in any city or place where a visitor feels a strong connection with the location where the tourist experience is being lived.

Finally, when tourists get a familiar feeling with a specific place, they may also develop a sense of “ownership” regarding the destination itself (Kumar and Nayak, 2019). In this perspective, the SOB to a destination also refers to the social connections between individuals tied to the social or community environment of the destination (Hammitt et al., 2004; Prayag and Ryan, 2012). Consequently, a person's SOB plays an important role in connecting him/her to the local social context, to the places and to the surrounding things (Hagerty et al., 1992).

Despite the relevance of this construct in the tourist field (Hung et al., 2019), it is surprising how the majority of studies have examined the SOB antecedents in the online consumer behavior context (e.g. Peng and Chen, 2020; Wiese and Akareem, 2019; Yang, 2019), especially by exploring its relationship with satisfaction (Chen et al., 2020; Zhou et al., 2019; Lin et al., 2014). Indeed, even if both SOB and satisfaction have been conceptualized as two emotional constructs, SOB can be considered as an outcome since it leads to stronger individuals' reactions with respect to satisfaction (Zhou et al., 2019). More in detail, different authors underlined how satisfied customers tend to become loyal ones (Kurniawati et al., 2021) by also transforming themselves into potential brand advocates (Sashi et al., 2019; Parrott et al., 2015), with consequently a marked sense of belongingness and ownership toward the satisfactory brand, service, firm, or entity (Machado et al., 2014).

By focusing on the tourist destination context, few authors (Peng and Chen, 2020; Lee et al., 2012) have investigated the role of satisfaction as antecedent of SOB, especially with respect to a cultural destination. More in detail, only Hung et al. (2019) hypothesized this relationship in the cultural domain by corroborating the positive impact of satisfaction on tourists' SOB in cultural destinations. Therefore, the first research hypothesis is formulated as follows:

The extant literature (Hung et al., 2019; Hagerty et al., 1992) underscored the connection between SOB and the involvement of people in experiences and, in tourism, the experiential perspective has now taken on a well-consolidated role in the staging of experiences that can engage tourists at a destination (Pencarelli and Forlani, 2018; Pine and Gilmore, 1999). The participation and involvement in quality activities (Kusumawati and Rahayu, 2020) create positive experiences and increased value for participants (Castellani et al., 2020), on the assumption that a higher level of experience-involvement will lead to memorable and authentic experiences for the consumer (Salla Sallaku and Vigolo, 2022; Zatori et al., 2018).

In striving to understand the types of experiences that can have a positive impact on tourists' SOB, prior studies have taken a close look at in loco activities (Lu et al., 2015). More recently, Hung et al. (2019) incorporated the variable “on-site activity involvement” in the Mehrabian-Russell model (M-R model), identifying the moderating role this variable plays in the relationship between tourists' satisfaction and SOB. In cultural destinations, where tourists have left their customary habitat and are motivated to temporarily immerse themselves in the culture they are visiting (Cetin and Bilgihan, 2016), it can be hypothesized that participation and involvement in on-site activities can positively affect tourists' satisfaction vs SOB relationship. Consequently, the following hypothesis is made:

Although previous research examined on-site activity involvement (e.g. Salla Sallaku and Vigolo, 2022; Kusumawati and Rahayu, 2020; Hung et al., 2019; Zatori et al., 2018), a lower number of studies have focused on the analysis of its social media dimension (Amaro and Duarte, 2015). Indeed, even if the investigation of the users' real-time involvement in social media is currently gaining more and more attention (e.g. Raji et al., 2019), the majority of the extant literature is focused on sectors other than tourism, such as the automotive (Raji et al., 2019), publishing (e.g. Naskar et al., 2021), social commerce (e.g. Alsoud et al., 2022) and food and beverage industry (e.g. Gastaldello et al., 2022).

By concentrating on the tourism field, existing research has mainly focused on the social media role in the planning phase of the holiday (Pop et al., 2022; Tuclea et al., 2020), without deepening the analysis of their real-time functions when the tourist is on-site (Cassia et al., 2021). However, a growing number of studies have recently begun to highlight the relevance of social media's real-time role in the communication and marketing strategies of tourist attractions, territories, cities, destinations and relative organizations (e.g. DMOs) (Tugores-Ques and Bonilla-Quijada, 2023; Bonilla-Quijada et al., 2021; Capolupo et al., 2020; Molinillo et al., 2018).

In this vein, different studies emphasized how the new media platforms have made possible new types of connectivity and involvement of the tourist at a given destination (Amaro and Duarte, 2015; Munar and Jacobsen, 2014). For instance, in a study on the authenticity of a destination, Kim and Kim (2020) demonstrated that when there is an emotional attachment to a destination, tourists are more likely to communicate their engagement through social media.

Subsequently, Zollo et al. (2022) underscored how social media platforms could significantly contribute to the competitive advantage of cultural places by also improving visitors' level of satisfaction and attracting new tourists.

However, despite these studies, a paucity of research specifically focused on the analysis of the social media role in the real-time involvement of cultural tourists emerges (Pasquinelli et al., 2022; Zollo et al., 2022; Campillo-Alhama and Martínez-Sala, 2019).

Based on these assumptions, the present study posits that social media involvement can positively moderate the relationship between tourists' satisfaction and their SOB in the cultural destinations. Thus, the third hypothesis is formulated as follows:

When visitors feel a SOB to the destination they tend to align their objectives with those of the local hosting community, also showing a willingness to do something in terms of the behavior(s) they display (Hahm et al., 2018). Tourists' SOB can therefore influence their behavioral intentions (Jones et al., 2000.) In this vein, some studies involving brand communities and brand identification also found a positive correlation between SOB and certain behavioral intentions, such as recommending a destination or intending to return (Cheung and Lee, 2012).

Hung et al. (2019) showed how a tourist is more likely to return to and recommend to others a given cultural destination when he/she feels a strong SOB to the destination itself. In two different studies, Han et al. (2019a, b) investigated the moderating influence of SOB. In the first one, in a context of community-based tourism, they found that it positively influenced both word of mouth (recommendation) and the intention to return; in the second one, in a context of halal tourism, they revealed how post-purchasing decisions made by Muslim clients were clearly moderated by their SOB to the tourist destination.

Moreover, SOB can significantly contribute to fostering positive conducts of tourists in loco (Kumar and Nayak, 2019). These can entail typical, good citizen actions that help to protect the location (Han et al., 2019b), such as sustainable behaviors (Cheng et al., 2017).

Among these, Environmentally Responsible Behavior (ERB) assumes a key role in contributing to a sustainable development of tourism in destinations (Alazaizeh et al., 2019; Chiu et al., 2014) since it “refers to a series of behaviors carried out by tourists, aiming at reducing negative environmental impacts and advocating sustainable resource use at destinations” (Luo et al., 2020, p. 2). Examples include recycling, picking up litter, being respectful of the local flora and fauna (Su et al., 2020), engaging in social activities like local heritage conservation, or enhancing and respecting the local culture and population (Blancas et al., 2016). In particular, these actions could be encouraged through the direct and constant interaction between tourists and DMOs which, as evidenced by the literature (Leung and Jiang, 2018), can play a key role in the tourists' SOB creation through the building of long-lasting relationships (Băcilă et al., 2022).

A study conducted by Alonso-Vasquez et al. (2019), focused on place attachment connected to events, revealed how positive interactions between participants and organizers fostered the visitors' SOB. This, in turn, leads to the adoption of a pro-environment attitude during the course of the festival, making them sustainable.

Similarly, in our study, a positive tie between SOB and ERB of tourists has been posited. Thus:

Research focused on environmentally sustainable behaviors is paying more and more attention to the analysis of their spillover effects. Recent studies (e.g. Wu et al., 2021; Xu et al., 2020) investigated the influence of ERB on additional related behaviors. In this context, some studies corroborated a positive effect (i.e. positive spillover effects) (e.g. Ye et al., 2022; Nash et al., 2019), while others a negative one (i.e. negative spillover effect) (e.g. Thomas et al., 2016; Truelove et al., 2014). Despite contradictory findings, in the present study, we hypothesize a positive influence of ERB on a specific spillover effect, namely Pro-Environmental Behavior (PEB). Indeed, although some scholars considered these two behaviors as interchangeable constructs (Lee et al., 2013), they represent two different typologies of attitudes. Notably, while ERB specifically refers to tourists' on-site behaviors (Su et al., 2020), PEB is related to future attitudes they might assume during upcoming holidays (Wang and Zhang, 2020; Li and Wu, 2019). Thus, we deemed it interesting to investigate, in the specific context of cultural destinations, whether ERB can have an impact on visitors' future pro-environmental behaviors.

Drawing from the research by Nilsson et al. (2017) - which underscores how specific behaviors can positively promote initiatives in favor of environmental conservation – and from the growing evidence that, in certain circumstances, specific behaviors “can affect involvement in other actions aligned with the same goal” (Nash et al., 2017, p. 2), we assume that ERB can have a positive impact on PEB. Thus, the hypothesis is formulated as follows:

Lastly, although previous studies explored the relationship between ERB and loyalty to the destination, the findings appear contradictory. For instance, Azinuddin et al. (2022) surprisingly revealed that tourists' pro-environmental behavior does not influence their loyalty toward the tourist destination. Collaterally, as held by Huang and Liu (2017), some studies showed that the behavioral intentions of tourists are related to their worries, their environmental awareness and their choice of destination (Mihalic, 2013); furthermore, these intentions can also impact their willingness to recommend a given destination (Altunel and Erkurt, 2015).

In the wake of these studies, we put forth the hypothesis that the adoption of ERB at a destination has a positive effect on the intention to recommend the destination itself, formulated as follows:

To test the hypotheses, a survey has been carried out. In particular, questionnaires were administered to tourists in the city of Urbino, that has been characterized, in the last decade, by a significant tourist growth, interrupted during the pandemic and recovered in the post-COVID-19 period (Assoturismo-Confesercenti, 2022).

Urbino has been selected for several reasons: (1) its cultural and international tourist dimension; indeed, it is an important UNESCO World Heritage site, famous throughout the world for being one of the main capitals of the Italian Renaissance as well as the birthplace of Raffaello Sanzio; (2) its strategic relevance in terms of municipality with a historical, artistic and landscape vocation also with respect to the main European cultural cities (XXIV Rapporto sul Turismo Italiano, 2019-2020); (3) its urban morphology, that is represented by a limited Renaissance village, where it is impossible to separate the residential dimension from the tourist one, thus constituting an immersive experience for the tourist.

The questionnaire has been administered between June and July 2022, which represents the period in which the main tourist flows of the city are concentrated, thus offering a greater probability of inclusion, in the sample, of visitors from various Italian regions and foreign countries.

The respondents were recruited as part of a student project of the Faculty of Languages of the University of Urbino; they were handed a digital tablet with the questionnaire on it, which they could complete out on their own. On the tablet, the questionnaire could be opened and filled out using “Google Forms,” granting a digital approach that allowed for the elimination of printed paper and speedier processing of data that was automatically stored in an online database and then analyzed. A translation-back-translation method was employed to carry out the survey in multiple languages.

The study used a non-random sampling approach. In particular, tourists have been firstly approached with the final aim of verifying some specific conditions: that the tourists (1) had completed their visit to the city; (2) were over 18 years old. After verifying these conditions, the interviewers explained both the purpose of the study and the data processing regulations (Regulation (EU) 2016/679 GDPR). When filling out the questionnaire, the interviewers assumed a passive attitude to avoid influencing the respondents, remaining available for any clarifications.

After eliminating incomplete responses and discarding respondents with a uniform response style (Völckner et al., 2010), we ended up with 647 valid responses. The sample consisted of 52% female and 47% male respondents and 1% who preferred not to respond (for Gender). Of the total respondents, 14% were between 18 and 29 years of age, 39% were between 30 and 49 and 47% were above 50. In total, 60% of the respondents were Italians (coming from 18 Regions), while 40% were foreigners (from 24 different countries).

With a total of 647 respondents, the sample is above the rule of 200 and the sample to item ratio is 29, which is more than the acceptable ratio of 5:1 (Gorsuch, 1983). Thus, an adequate sample size was achieved. Kaiser-Meyer-Olkin (KMO) as well as Bartlett's Test of Sphericity to measure sampling adequacy were calculated. KMO is 0.831 (> than 0.5) and Bartlett's Test of Sphericity is significant at 0.000 (below p < 0.05); therefore, both values are over the threshold and the data is suitable for factor analysis.

For the operationalization of the constructs, existing and empirically validated scales have been selected from established contributions and then adapted to suit the present research field. More in detail, concerning “satisfaction” and “sense of belonging”, we have adopted the scales of Hung et al. (2019), who, in turn, have extracted them from the studies of Taplin (2013) and Zhao et al. (2012), specifically focused on tourist satisfaction within two different contexts (i.e. a metropolitan zoo and virtual communities). Notably, we have proceeded by adopting the scales of Hung et al. (2019) since the authors have analyzed the same research area of this study (i.e. cultural tourism destinations). Moreover, to ensure the scales goodness, we have also realized a literature review process that has allowed to identify different studies (e.g. Peng and Chen, 2020), which have adopted and then adapted the Hung et al. (2019) scales to their specific context of analysis. Regarding “pro-environmental behavior”, “Environmentally responsible behavior” and “recommendation intention”, the scales of Wang and Zhang (2020), Su et al. (2020) and Altunel and Erkurt (2015) have been respectively adopted and applied to the destination context. With respect to the moderators, for the “social media involvement” construct, 4 items from Amaro and Duarte (2015) have been selected since these authors were the first to propose a measurement for the tourists' level of involvement in social media. Lastly, for “on-site activity involvement”, the scale of Hung et al. (2019) has been chosen since, similarly to “satisfaction” and “sense of belonging”, it has been adopted within the context of cultural tourism destinations. Overall, survey respondents were asked to indicate their level of agreement for each of the items using a seven-point Likert scale, from totally disagree (1) to totally agree (7).  Appendix contains the complete list of the items and the source adopted for each construct.

Several analyses were conducted to test our model. Exploratory factor analysis, confirmatory factor analysis and structural equation modeling were used to test the hypotheses. Employing principal factor analysis showed that all items loaded on the proposed constructs. Overall, the seven factors explain a 68.6% cumulative variance. None of the 22 items had significant cross-loadings (>0.50). All scales are reliable with Cronbach's alpha values higher than 0.7 (see Table 1).

Common method bias is assessed by Harman's single-factor test. This method assumes that the presence of common method variance is indicated by the emergence of either a single factor or a general factor accounting for the majority of covariance among measures (Podsakoff et al., 2003). The single factor results in explaining 25.9% of the variance. Given that this value is far below the 50% mark, we are confident that common method bias is not an issue with our data. Further, the order of the questions was chosen carefully. Participants were also informed about the anonymity and confidentiality of the survey, as well as the fact that there are no “right” or “wrong” answers and that they should answer as honestly as possible.

To assess the multicollinearity, a series of regression models were run on the various constructs to calculate the variance inflation factor (VIF), and the tolerance test for multicollinearity was conducted (Kleinbaum et al., 1988). The values for the VIF are between 1.01 and 1.21 and the tolerance test's values are between 0.83 and 0.95; therefore, no evidence for multicollinearity exists. Further, the convergent and discriminant validity of the constructs was assessed through a confirmatory factor analysis. Average variance extracted (AVE) and composite reliability (CR) form convergent validity. To obtain convergent and discriminant validity, the AVE should be > 0.40 (Floyd and Widaman, 1995) and the CR should be > 0.60 (Bagozzi and Yi, 1988). We found the AVE values to be between 0.616 (social media involvement) and 0.858 (recommendation intention) and CR values to range between 0.783 (environmentally responsible behavior) and 0.948 (recommendation intention). Thus, all the AVE and CR values are acceptable. Discriminant validity was established by comparing the square root AVE values (in the diagonal) with the correlation estimates. For discriminate validity, the diagonal elements should be larger than the off-diagonal elements. The details concerning the mean and standard deviation of the constructs, as well as Cronbach's α, AVE, CR and correlation values, are displayed in Table 1, below.

Structural equation modeling (SEM) was employed using SPSS AMOS 26 to test the hypotheses for the present study. For testing the interaction hypothesis, we adopted the double-mean-centering strategy (Lin et al., 2010), i.e. all the observed indicators are mean-centered before creating the product terms, and the product terms are then mean-centered before fitting the model with the latent interaction factor. Lin et al. (2010) noted that this approach improves modeling latent variable interactions compared to previous approaches that required incorporating the use of mean structures and non-linear constraints (see also discussion by Steinmetz et al., 2011). They argued that the double-mean centering strategy is preferable to using residual centering (orthogonalizing) on two grounds: (1) the residual centering strategy is potentially more cumbersome given it requires a two-step process for centering; (2) although residual centering and double-mean centering will result in equivalent outcomes when the first order factors are bivariate normal, the latter approach performs better when the factors are not normal. Further, we utilized an “all-pairs” approach when forming product indicators (Foldnes and Hagtvet, 2014). Then, the interaction is probed by generating simple slopes of the effect of a focal independent variable at different levels of a moderator variable and testing them for statistical significance. The following formula is applied: $SS=(β1+β3M)$⁠, where M represents the level of the moderator a given test is being carried out at. The simple slopes are tested using the AMOS bootstrap utility. To determine levels of the moderator simple slopes are generated at estimates of the standard deviation of the moderator (Hayes, 2013). In the presence of a significant interaction, simple slope analyses were estimated to evaluate the effect of satisfaction at high (+1 SD) mean, and low (−1 SD) values of on-site activity involvement, as well as the effect of satisfaction at high (+1 SD), mean and low (−1 SD) levels of social media involvement, thus enabling statistical comparisons (Gaudreau et al., 2016).

The findings show an acceptable model fit with χ2 = 588.275; df = 144; p < 0.001; χ²/df = 4.085; NFI = 0.931; RFI = 0.918; IFI = 0.947; TLI = 0.936; CFI = 0.946; and RMSEA = 0.069. SOB is influenced by satisfaction (β = 0.178; p < 0.001), confirming H1. SOB (β = 0.407; p < 0.001) influences environmentally responsible behavior, thus confirming H4. Environmentally responsible behavior influences pro-environmental behavior (β = 0.377; p < 0.001), supporting H5, and recommendation intention (β = 0.290; p < 0.001) supporting H6. Figure 1 provides an overview of the significant results of our model testing, schematically represented.

Lastly, we tested to see if on-site activity involvement and social media involvement moderate the relationship between satisfaction and SOB. To test these two hypotheses, we created interaction terms for satisfaction and on-site activity involvement as well as for satisfaction and social-media involvement.

The results of the moderation tests revealed that the relationship between satisfaction and SOB is significantly strengthened when tourists' on-site activity involvement increases, supporting H2. No such moderating effect was found for social media involvement, thus H3 was not confirmed. The results are displayed in Table 2, below.

We conducted a simple slope analysis for the significant interaction effect of satisfaction with on-site activity involvement. Simple slope analysis showed that when on-site activity involvement is medium (at the mean; B = 0.180; p = 0.001) or high (1 SD above the mean; B = 0.123; p = 0.009), the relationship between satisfaction and SOB is higher compared to the relationship between satisfaction and SOB when on-site activity involvement is low (1 SD below the mean; B = 0.057; p = 0.006). Thus, the more tourists were involved with the location and the higher their satisfaction was, the more they felt a SOB to the location.

Promoting tourists' SOB represents a fundamental strategy for the long-term success of cultural destinations, in terms of both competitiveness and sustainability. Indeed, one of the main findings of the study confirms the positive connection between tourists' satisfaction and the ability to develop a SOB to cultural destinations (Hung et al., 2019). Moreover, our research offers important points for reflection on the experiential factors that can influence the relationship between satisfaction and SOB. On the one hand, the study confirms the moderating role played by on-site activities in the cultural destinations' context (Cetin and Bilgihan, 2016); on the other hand, the findings reject the hypothesis of tourists' social media involvement as a moderator in the relationship between their satisfaction and SOB. Our research represents one of the first attempts to deeply examine the adoption of these tools, given the widespread interest shown by travelers as well as scholars in the field (Pop et al., 2022; Chen et al., 2022; Wong et al., 2020). The fact that this hypothesis is not supported opens the door for reflections on two fronts that can be useful for future studies. First, the SOB construct also refers to social connections between and among visitors and the destination community (Prayag and Ryan, 2012; Hammitt et al., 2004); consequently, in those on-site experiences that present key factors for social interaction and local authentic clues (Cetin and Bilgihan, 2016), tourists will recognize authentic activities that can connect them to the local social context, to the people and to the places, in spite of on-line tools. Second, while extant research tends to focus on the analysis of the social media role during the consumer decision process phase or journey (Sallaku and Vigolo, 2022; Pop et al., 2022; Tuclea et al., 2020), the present study underlines the key relevance of specific investments in both social media platforms' management and specific functions during the tourist's holiday, in the perspective of “Digital Tourist Journey” (Guerreiro et al., 2019).

Indeed, given that social media can be an extremely important tool during a tourist's entire stay, tourist organizations could focus their attention on the quality of these platforms, especially in terms of availability of information and interaction with the visitor (Kim and Kim, 2020; Narangajavana et al., 2017).

Furthermore, another aspect is related to the fact that the rapid and significant growth of cultural tourism is creating increasingly serious environmental problems, which are drawing more and more attention to tourists' behaviors (Cheng et al., 2017). In this vein, the present study unveils how tourists' SOB can trigger a positive effect on their ERB in cultural destinations (Alonso-Vasquez et al., 2019), thus confirming how emotional constructs can crucially affect sustainable behavior (Su et al., 2020).

Moreover, we also explore a further debated aspect related to the spillover effects of tourists' responsible behavior (Wu et al., 2021; Xu et al., 2020). In particular, our findings confirm the positive influence of ERB, both on future pro-environmental attitudes of tourists and on their intention to recommend the destination to others (Nash et al., 2017; Huang and Liu, 2017; Altunel and Erkurt, 2015), thus triggering a virtuous process in the formation of responsible and loyal visitors.

From a theoretical point of view, the study contributes to filling some gaps in the more recent literature focused on the behaviors and attitudes of travelers in the cultural tourism domain. In particular, six main areas of interest have been investigated: (1) the current debate on the 2030 agenda's goals, especially by focusing on the relationship between social, cultural, environmental sustainability and tourists' behaviors (Alonso-Muñoz et al., 2023); (2) the relationship between satisfaction and SOB, thus enriching the extant literature mainly centered on the first construct (Hung et al., 2019; Lin et al., 2014); (3) the analysis of SOB in the context of touristic experiences by expanding the body of previous studies that primarily associated this construct with brands/products (Hung et al., 2019); (4) the moderating role of tourists' on-site and social media involvement, which allows to bridge an additional gap related to the scarcity of studies focused on these two constructs in the cultural tourism context (Amaro and Duarte, 2015); (5) the relationship between SOB and ERB, thereby contributing to the stream literature of sustainable behaviors (Lin and Lee, 2020); and (6) the analysis of the spillover effects of ERB in the context of cultural tourism (Xu et al., 2020).

At the managerial level, the present study offers some important implications for different actors in the tourism sector and specifically in the cultural one. In detail, this study is relevant for policy makers, destination marketing organizations (DMOs), industry associations and tourism operators by showing how important it is for them to enhance tourists' SOB through their constant on-site involvement at the cultural destination, especially during specific sustainable activities.

Policy Makers and DMOs, as destination governance entities, play a crucial role in pushing tourism operators to qualify the involvement-experiential activities, especially by promoting sustainable on-site actions (e.g. recycling, adoption of eco-friendly transportation, renewable energy production in public places) able to favor, also in the light of the Agenda 2030, the adoption of correct sustainable practices within the destination (Alonso-Muñoz et al., 2023). Furthermore, destination government subjects should also implement, especially in cultural destinations, effective systems and metrics aimed at monitoring and assessing the sustainable performance of the destination (Zhang et al., 2023).

Even individual tourism businesses, which represent the backbone of the tourism sector (Dredge et al., 2019), must enhance specific on-site touristic activities that can make the destination memorable and unique (Zatori et al., 2018), also through the adoption of sustainable models of consumption and production.

The realization of sustainable activities would also consist in the exploitation of the communication effects by also planning “sustainability-contents SEO”, which allows to develop and maintain, in the long term, trusting relationships between companies and publics (Confetto and Covucci, 2021).

These activities can be a starting point to build loyalty processes (Huang and Liu, 2017) for tourists who are currently paying more and more attention to the experiential dimension and eco-awareness tied to sustainability concerns. Therefore, DMOs and local businesses need to set up quality experiences (Kusumawati and Rahayu, 2020) that entail involvement in authentic realities, offering sustainable living spaces that go beyond the classic touristic pathways. This requires the involvement and commitment of volunteers and the local community, because SOB to and within a destination also depends on the social connections that are forged among the members of the surrounding community (Prayag and Ryan, 2012).

Fostering tourists' SOB will also affect their ERBs, which are fundamental for minimizing the impact on the destination and guarantying social equilibrium with the local community. This equilibrium will also allow to perceive tourism – not as an invasive phenomenon – but as one that can build up the economic, social and environmental value of the destination itself (Blancas et al., 2016).

Potentially, the adoption of sustainable behaviors within a destination should inspire DMOs' communication and marketing strategies, so as to also attract qualified segments like responsible tourists (Jopp et al., 2022).

Finally, the SOB could be developed through multiple physical, digital and phygital touchpoints during the different customer journey phases, and especially when the tourist is living the destination experience (Sallaku and Vigolo, 2022). Given that our results show that, currently, social media do not play a role in the relationship between satisfaction and SOB, this should motivate destinations and DMOs to improve their social media tools in order to catch all the opportunities for interaction during a tourist's stay (Molinillo et al., 2018). This requires that the human capital employed at the destination must be sufficiently trained, especially for encouraging tourists to share content online (Munar and Jacobsen, 2014).

A few limitations of this study provide opportunities for future research.

First, given that this study was limited to data collected from a single cultural heritage site, the city of Urbino, future research should consider different typologies of destinations.

Second, as regards tourists' experience-involvement, specific typologies of on-site cultural experiences should be analyzed to determine which ones have the greatest influence on SOB (e.g. guided tours, technology-driven experiences, relational experiences with the local population, cultural and enogastronomic experiences, etc.). In a similar vein, it could be interesting, in future studies, to investigate additional outcomes of SOB, such as destination resilience and competitiveness (Dwyer et al., 2019).

Third, the role played by social media could be further examined through qualitative studies regarding both demand and supply to delve into this underexplored theme (Sallaku and Vigolo, 2022).

Moreover, to address the issue of discordant literature in this field, future research should endeavor to confirm the relationship between visitors' SOB and their sustainable behavior(s) by also adopting other conceptualizations of ERB constructs that take into account the various dimensions of sustainability (Woosnam and Ribeiro, 2023) and in different cultural contexts (Tu and Ma, 2022).

Finally, it would be worthwhile to assess the impact of behaviors in the aftermath of tourist quality experiences (Kusumawati and Rahayu, 2020), both positive and negative (Su et al., 2020), as well as the relationship between constructs, taking into account the different tourists' backgrounds.

The authors thank, among the authors, a researcher awarded with a fixed-term type-A research contract on innovation topics as per art. 24, para. 3, of Italian Law no. 240 of 30 December 2010, co-financed by the European Union - NOP Research and Innovation 2014-2020 resources as per Italian MD (No: 1062) of 10 August 2021.

Since acceptance of this article, the following author(s) have updated their affiliation(s): Sabrina Hegner is at the School of International Business, City University of Applied Sciences Bremen, Bremen, Germany.

Table A1